A general Laguerre transform and a related distance between probability measures (Q1081068)

The authors are motivated to introduce, in the first instance, the direct sharp Laguerre transform \(f(x)\leftrightarrow (f_ n^{\#})^{\infty}_{-\infty}\), since in certain statistical and applied probabilistic problems, where the fundamental \(L_ 1\) character of probability densities is present and the \(L_ 2\) property is absent, the Laguerre dagger transform \(f(x)\leftrightarrow (f_ n^{\dag})^{\infty}_{-\infty}\) does not exist in \(l_ 2\). They define the direct Laguerre sharp transform by the equation: \(f_ n^{\#}=\int^{\infty}_{-\infty}\{(h_ 0*f)(x)\}h_ n(x)dx\), where \((h_ 0*f)(x)=\int^{\infty}_{-\infty}h_ 0(x-y)f(y)dy\in L_ 2.\) The spaces \(L_ 1\) and \(L_ 2\) are defined in the usual way. It is shown that the linear function space \(L_ 2\) defined by \(L_ 2^{\#}=\{f(x): R\to R|\| f\|_ 2^{\#}<\infty \}\), where \[ \| f\|_ 2^{\#}=\sqrt{\int^{\infty}_{-\infty}\{(h_ 0*f)(x)\}^ 2dx}=\sqrt{\sum^{\infty}_{n=-\infty}f_ n^{\#2}}, \] is a Hilbert space and contains both \(L_ 1\) and \(L_ 2\). They then discuss the direct Laguerre sharp transform for a probability measure \(\mu\) on R. Some properties of the first differences \(\Delta h_ n\) of the extended Laguerre functions, which are needed in the investigations, are derived, including the asymptotic behaviour as \(n\to \infty\). It is further shown that the weak convergence of a sequence of probability measures is equivalent to the convergence in the Laguerre sharp norm. Lastly several interesting properties of the Laguerre sharp norm associated with the space \(L_ 2^{\#}\) are derived and in the sequel convolution properties of the Laguerre sharp norm lead to some useful inequalities.